Highly mac-tropic HIV-1 R5 Envs that efficiently bind CD4 are associated with HIV neurological disease including HIV associated dementia (HAD). However, our knowledge of the properties and determinants of Envs that colonize brain and evolve into highly mac-tropic, neurovirulent variants are incomplete. Without this, we will not fully understand how and where neurotropic HIV-1 strains arise and develop into neurovirulent forms. Our preliminary data shows that R5 Envs derived from brain tissue of subjects without or with only minor neurocognitive disease (N/MCs) were mainly non-mac-tropic compared to those in HAD brain. This data suggests that mac-tropism is not required for HIV to enter brain tissue and opens up the possibility that HIV is carried in by infected T-cells, before mac-tropic and neurovirulent variants evolve. Env determinants that confer mac-tropism in HAD include residues within or proximal to the CD4bs. However, a single universal Env determinant of mac-tropism has not been identified and residues in different parts of gp120 e.g. the V1V2 and V3 loops were shown to modulate mac-tropism. These residues presumably increase the affinity of the Env trimer for CD4 either (1) directly via changes in contact residues or (2) indirectly by improving access to CD4 or enhancing CD4-induced conformational changes. We recently established a trimeric Env: CD4 binding assay, which shows mac-tropic Env trimers from brain bind CD4 efficiently, while non-mac-tropic Envs don't. We will use this and other assays to investigate how Env conformation and function varies in immune and brain tissue of N/MC and HAD subjects. Our hypothesis is that HAD is determined by the evolution of Envs with a higher affinity for CD4, that (1) carry less tightly closed Env trimers with a more exposed CD4bs, (2) are more easily triggered by CD4 and (3) mediate efficient macrophage infection. We propose 3 aims: Aim 1: To analyze HIV-1 replication and tropism in immune and brain tissue of N/MC subjects. We will establish whether envs recovered from brain of N/MC subjects are derived from actively replicating virus and the extent they are mac-tropic. Aim 2: To establish how Env conformation, tropism, receptor interactions and neurotoxicity vary in N/MC and HAD subjects. We will investigate how Env trimer conformation and function vary between immune and brain tissue of N/MC and HAD subjects. Aim 3: To identify determinants in gp120 and gp41 that modulate mac-tropism and evaluate their impact on Env conformation and function. We will identify Env determinants and properties associated with (1) colonization and viral replication in brain tissue and (2) neurovirulence and neuroAIDS. We will provide new insights to help develop strategies to identify HIV+ subjects at risk of neuroAIDS and `cure' approaches to attack persistent HIV in the CNS.